The incidence of cutaneous malignant melanoma (CMM) is rapidly rising among Caucasians and is estimated to affect approximately one of every 100 Caucasian Americans within their lifetime. Melanoma results from the oncogenic transformation of the pigment-producing cells of the skin and hair, melanocytes. (Rhodes, et al., JAMA, 258:3146-3154 (1987); Jimbow, et al., Physiology, Biochemistry and Molecular Biology of the Skin, Vol. II, ed. Goldsmith, (N.Y.: Oxford Univ. Press, 1991); Elder and Clark, Pigment Cell, 8:51-80 (1987)). Although CMM is easily recognized clinically and treated surgically, this form of skin cancer causes about 6700 deaths each year in the USA, due to the propensity for melanoma to metastasize (NIH Consensus Development Panel on Early Melanoma, JAMA 268:1314-1319 (1992)), thus making it the most serious form of skin cancer. The risk factors for cutaneous melanoma include increased age, race, familial occurrence of this disease, ultraviolet radiation (UVR) exposure, dysplastic nevi, fair skin, among others. (Rhodes, et al. (1987); Jimbow, et al., (1991); Elder and Clark (1987); NIH, et al. (1992); Elwood and Lee., Sem. Oncol., 2:149-154 (1975)). Conventional chemotherapy strategies employing cytotoxic or anti-proliferative agents have not been highly successful for the treatment of metastatic melanoma to date. Therefore, oncologists are in need of more effective and safe therapies for the treatment of melanoma.
Several laboratory animal models have been developed to study various aspects of melanoma tumorigenesis, genetics, immunology and therapy (Dooley, Oncol. Res., 6:1-9 (1994)). Although no single animal model perfectly matches the genetic, biochemical, and pathological characteristics of human melanoma, each of the common models has some value in specific research areas for comparative studies relative to human melanoma. The established mammalian models, i.e. rodents (primarily mouse) and pigs (Sinclair swine) have some value in specific areas for comparative studies of human melanoma, but they do not provide a satisfactory match for the genetic, biochemical and pathological characteristics of the human disease.
Perhaps the most widely used mammalian model is mouse, although other rodent species have been used to a lesser degree. Mice derive their usefulness from the abundance of genetically-defined inbred strains. The minimal genetic heterogeneity within inbred strains permits cancer researchers to overcome many of the problems inherent in tumor transplantation studies, as both syngeneic and immunoincompetent strains (e.g., athymic nude mice) are available. (Kruger and Pershing, Pharmacology of the Skin, ed. Mukhtar (London: CRC Press, 1992); The Nude Mouse in Oncology Research, ed. Boven and Winograd (London: CRC Press, 1994); Dooley, et al., Lab. Animals Sci., 43:48-57 (1993)). Mice have been of great utility in the development of a variety of melanocytic cell lines exhibiting varying degrees of tumorigenicity and metastatic potential. (Kruger, et al, (1992); Bennett, et al., Int. J. Cancer, 39:414 (1987); Dooley, et al., Oncogene 3:531-536 (1988); Wilson, et al., Cancer Res. 49:711-716 (1989); Fidler, Nature New Biol. 245:148-149 (1973); Fidler, et al., J. Natl. Cancer Inst. 67:947-956 (1981)). Furthermore, immunoincompetent strains, such as nude and skid mice, are permissive for growth of melanoma cell lines from murine sources.
In some inbred mouse strains, treatment of shaved skin with a combination of chemical carcinogens (e.g., DMBA, croton oil, TPA, etc.) or a chemical carcinogen plus UVR produces melanocytic lesions. (Donawho, et al., J. Immunother, 12:187-193 (1992); Yuspa and Dlugosz, Physiology, Biochemistry and Molecular Biology of the Skin, 2d ed., ed. Goldsmith (N.Y.: Oxford Univ. Press, 1991); Bickers and Lowy, J. Invest. Dermatol. 92:121s-131s (1989)). However, treatment of murine skin by a single agent alone seldom or never produces any of these melanocytic lesions.
In the past few years, transgenic mouse models have been created that spontaneously develop melanomas. However, the tumors generally lack pigment. (Iwamoto, et al., The EMBO J. 10:3167-3175 (1991); Bradl., et al., Proc. Natl., Acad., Sci. 88:164-168 (1991); Larue, et al. Oncogene 8:523-531 (1993)). The transgenic melanoma models, utilizing the melanocyte-specific tyrosinase promoter fused to an activated oncogene, are showing significant promise for experimental studies of the cooperativity between known oncogenes and additional factors, such as UVR, tumor promoters, etc., but their potential utility in chemotherapeutic drug discovery efforts has not been examined.
A heritable form of congenital malignant melanoma is associated with Sinclair swine. This variety of pigs develops malignant melanomas in utero. The tumors exhibit rapid growth, invasion, and life-threatening metastases. (Dasgupta, et al., Pediatr. Dermatol. 6:289-299 (1989); Beattie, et al., Semin. Onco. 15:500-511 (1989); Green, et al., Cancer Genet. Cytogenet. 61:71-92 (1992)). Remarkably, some of these affected animals subsequently develop an immunological rejection of the tumors. In some instances, the anti-melanoma rejection process results in cutaneous vitiligo, presumably due to the ablation of normal skin melanocytes expressing antigenic determinants similar to the melanoma cells. This model is very promising both for the genetics of congenital melanoma induction and for the immunology of tumor rejection, but does not serve as an allogeneic grafting model.
With regard to experimental human melanoma studies, the availability of a large number of metastatic melanoma cell lines capable of growth in vitro has significantly advanced our understanding of melanoma biology. In addition, human metastatic cell lines have been grown by xenotransplantation in vivo in immunoincompetent nude mice, primarily for investigations of antineoplastic therapies. Human melanoma cell lines have also been of value in recent years for immunological studies, including the development of: (a) anti-melanoma antibodies directed against tumor-associated antigens (Carrel and Rimoldt, Eur. J. Cancer 29A:1903-1907 (1993)); (b) potential anti-tumor vaccines (Carrel, et al. (1993)); and (c) recently-conceived gene therapies. Human melanoma cell lines have also been used in oncogenesis studies.
The establishment of the laboratory opossum (Monodelphis domestica) as a useful melanoma model resulted from the dermatological and photobiological experiments of Dr. R. D. Ley and colleagues in Albuquerque, N. Mex. (Ley, Photochem. Photobiol. 46:223-227 (1987); Ley and Applegate, Models in Dermatology, pp. 265-275, ed. Maibach and Love (Basel: S. Karger, 1989)). One attractive aspect of the model is that, unique among the mammalian species so far examined, benign and malignant melanomas can be induced by chronic ultraviolet radiation exposure alone, without the concomitant requirement of application of chemical carcinogens. (Ley, et al., Photochem. Photobiol., 50:1-5 (1989); Kusewit, et al., Vet. Pathol., 28:55-56 (1991)). Pedigrees of a large Monodelphis colony maintained at the Southwest Foundation for Biomedical Research (SFBR) have been carefully documented since the first founders were imported in 1978. (VanOorschot, et al., Lab. Anim. Sci., 42:255-260 (1992)). Using the established sunlamp exposure protocol for exposing animals with suberythemal doses for up to 45 weeks, nevi were induced in 14% of those introduced to the protocol as adults. None of the nevi, however, progressed to metastatic melanoma. (VandeBerg, et al., Arch. Dermatol. Res. 286:12-17 (1994)). In a follow-up study, animals were introduced to the same protocol at the earlier weanling stage and two (2%) percent developed melanocytic nevi. A single animal developed malignant melanoma with presumptive metastasis to the spleen. (Robinson, et al., Arch. Dermatol. Res. 287:333-337 (1995)).
Some features of the established adult UVR exposure protocol impose limitations on its effective use. For example, it is time consuming, labor intensive, and therefore expensive. Most of the animals maintained on the protocol for up to one year are uninformative, and while the general state of health of exposed animals remains satisfactory, chronic exposure frequently leads to an undesirable side effect in the form of aggressive eye tumors. Furthermore, by the end of the protocol, animals are too old to mate, thus precluding their use in a breeding program.
The feature that most clearly distinguishes marsupials from other mammals is the immaturity of their young at birth. Neonate skin in Moiodelphis is thin, hairless, and not fully differentiated and is likely to absorb more ultraviolet radiation than adult skin. Monodelphis females, unlike the great majority of female marsupials do not possess a pouch so neonates are exposed on the mother's ventral surface. Thus, entire litters can be exposed to UVR and affected individuals can be used later for breeding. Furthermore, Monodelphis suckling young may serve as recipients for injection of tumor cells, due to underdeveloped immunosurveillance.